"Ideal Parent" Structure Learning for Continuous Variable Networks

TL;DR

This paper introduces a new method to efficiently learn the structure of Bayesian networks with continuous variables, including hidden variables, significantly reducing computational costs and enabling more complex models.

Contribution

The paper presents a general approach that speeds up structure search in continuous variable networks and allows efficient addition of hidden variables.

Findings

Method significantly reduces structure learning time.

Effective in both fully observable and hidden variable scenarios.

Demonstrated on multiple datasets.

Abstract

In recent years, there is a growing interest in learning Bayesian networks with continuous variables. Learning the structure of such networks is a computationally expensive procedure, which limits most applications to parameter learning. This problem is even more acute when learning networks with hidden variables. We present a general method for significantly speeding the structure search algorithm for continuous variable networks with common parametric distributions. Importantly, our method facilitates the addition of new hidden variables into the network structure efficiently. We demonstrate the method on several data sets, both for learning structure on fully observable data, and for introducing new hidden variables during structure search.